Studia Humana Volume 6:4 (2017) Contents Existence and Predication in Free Logics (Guilherme Kubiszeski)………………………………………………..…..........................................3 A Praxeological Approach to Intentional Action (Alan G. Futerman, Walter E. Block)……………..……………………….……...….......................10 Christianity, the Free Market, and Libertarianism (Christian Light, Walter E. Block)…………………………………………………....….…………34 The Quest to Solve Problems That Don’t Exist: Thought Artifacts in Contemporary Ontology (Bernardo Kastrup)………...................…………………………………………………………….45 Kennan and the Neglected Variable in Post–Socialist Societies: The Loss of Honest Dialogue and the Need for Empathy (Jesenko Tešan, Joan Davison)……………......................................................................................52 Adaptation in South Korean Society of North Korean Elite Defectors (Nicolas Levi)…………………………………………….……………………………………..…..62 Hindu Spirituality: How to Grasp the Divine? (Max Demtchenko, Andrew Schumann)…………………………………………………………….74 Book Review “Religion Explained? The Cognitive Science of Religion after Twenty-Five Years” (Hans Van Eyghen)…………………………………………………………………………….……79 Studia Humana Volume 6:4 (2017), pp. 3—9 DOI: 10.1515/sh-2017-0023 Existence and Predication in Free Logics Guilherme Kubiszeski Secretaria de Estado de Educação do Distrito Federal, Brasil email : [email protected] Abstract : This paper presents a fundamental difference between negative semantics for free logics and positive ones regarding the logical relations between existence and predication. We conclude that this difference is the key to understand why negative free logics are stronger, i.e., they prove more, than positive free logics. Keywords : free logics, existence, predication, square of opposition. 1. Introduction Classical first-order logic assumes that all singular terms in the domain of quantification denote. Free logics do not make such an assumption, as they allow the use of empty singular terms. This means that in free logics one can assign a truth-value not only to non-empty-termed atomic formulae, but also to empty-termed ones. The way that assignment is done is the precise difference between positive free logics and negative free logics. While in the latter all empty-termed atomic formulae are false, the former allow some of them to be true [4]. Therefore, the distinction is mainly semantic. It is important to note that “free” is a shorthand for “free of ontological commitment” regarding singular terms in a formal language. This means that not all singular terms in a free semantics must have an interpretation in a specific domain of “existing” objects. The emphasis on the formal character of the language is needed, since we do not want to consider Russell, for instance, a free logician. In other words, the acceptance of free singular terms in a natural language does not make one a proponent of free logics. A well-known inconvenient surrounding classical predicate logic is its inability to translate – using only existential quantification, individual terms and identity – sentences like “Socrates does not exist” without commitment to the existence of Socrates. The most intuitive way of translating the corresponding affirmative sentence is by means of the formula ‘∃x(x=s )’ (informally: “there is an x such that x is identical to Socrates”), which happens to be a valid formula in classical first- order systems, as every singular term must denote to make the sentence true – assuming, of course, that ‘ s’ is an element of the interpreted language. Since ‘~ ∃x(x=s )’ is false under all interpretations, those systems fail to provide an intuitive translation of the negative existential statements of everyday talk, such as “Sherlock Holmes does not exist” or “Santa Claus does not exist”. Of course, this incapacity does not rely upon a syntactic impossibility to symbolize non-empty terms, but rather on an assumed ontological commitment with every bound variable. Surely, there are syntactic ISSN 2299-0518 3 formulations of sentences including empty terms in non-free logics (see [5] and [6]), but such formulations resort to more than just existential quantification, individual terms and identity. On the other hand, free logics accommodate that kind of negative existential statements. Usually, free logics employ a one-place predicate ‘E!’, such that ‘ E!s’ is true if and only if ‘ s’ is non-empty. ‘ E!’ may be defined as follows: E!t =df ∃x(x=t ) If a term ‘s’ is non-denotative, ‘~ E!s’ (informally: “ s does not exist”) turns out true. This is the common ground between positive and negative free semantics: a formula asserting the existence of some object may be true or false depending on whether its singular term denotes. But when it comes to assign a truth value to atomic formulae of the form ‘ Pt 1…t n’, where ‘ P’ is an n-place predicate other than ‘ E!’ and ‘ t1’ …‘tn’ are singular terms, the situation changes. Positive free logics allow some formulae of the form ‘ Pt 1…t n’ to be true even when their singular term is empty; negative free logics, au contraire , assign the truth-value false to all formulae of the form ‘ Pt 1…t n’ when their singular term is empty. One could ask to what extent those negative semantics are properly free, since they do not seem to be suitable for treating some sentences containing empty singular names (for instance, fiction talk). However, it is important to note that they still allow individual constants and variables without denotation, and this is precisely what makes a logic “free”. Let us consider the semantical clauses for ‘ E!t’ and ‘Pt 1…t n’ in both families of free logics. 2. Semantics for Free Logics Assume a usual first-order language plus ‘ E!’ and call it L. Let < D, I> be a pair where D is a possibly empty set and I an interpretation function such that: (1) for each individual constant t of L, either I(t) ∈ D or t is undefined, and (2) for each n-place predicate P of L, I(P) ⊆ Dn. The pair <D, I> is a negative model for L. A valuation function V is a function from the set of formulae of L to the set {T, F} of truth-values. Formulae of the form ‘ Pt 1…t n’ and ‘ E!t’ are evaluated as follows [4]: • V(Pt 1…t n) = T if and only if I(t1), …, I(tn) ∈ D and < I(t1), …, I(tn)> ∈ I(P). • V(E!t) = T if and only if I(t) ∈ D. Informally, the first clause says that a predication or a relational sentence may be true only if its terms denote. So “Sherlock Holmes is a detective” and “Santa Claus wears a beard” turn out false. The second only says that an existential sentence like “Socrates exists” is true only if the term “Socrates” is non-empty. Positive semantics, on the other hand, are often split -domain semantics. Assume the same first-order language L. Then take a domain and split it into two: Di (the inner domain, 1 possibly empty) and Do (the outer domain, necessarily non-empty). I is an interpretative function such that for each individual constant t of L, either I(t) ∈Di or I(t) ∈Do. The triple < Di, Do, I> is a positive model for L. A valuation function V is a function from the set of formulae of L to the set {T, F} of truth-values. Formulae of the form ‘ Pt 1…t n’ and ‘ E!t’ are evaluated as follows [4]: • V(Pt 1…t n) = T if and only if <I(t1), …, I(tn)> ∈ I(P). • V(E!t) = T if and only if I(t) ∈ Di. 4 So positive free logics have a weaker clause for the truth of formulae of the form ‘ Pt 1…t n’ than negative ones, since they can be true even if none of their terms refer. The second clause, informally read, simply means that existence is the same as being in the inner domain. Informally put, “Sherlock Holmes is a detective” and “Santa Claus wears a beard ” may turn out true according to positive semantics. Members of the outer domain, understood as “non- existing” objects, may belong to the interpretation of one-place predicates, as well as be an element of n-tuples that form the interpretation of n-place predicates (for n > 1). 3. Lambert’s Square Lambert [3, p. 135] argued that the classical principles of Specification and Particularization give rise to a square of opposition built not only out of quantified formulae, but out of non- quantified ones as well. It looks like this: Lambert correctly showed that the inferences from ∀xPx to Ps and from ~Ps to ∃x~Px only works with an important presupposition – that ‘ s’ denotes, or equivalently, that ‘ ∃x(x=s )’ is true. He chose to consider the classical version of both principles simply wrong and to weaken them as follows: • Specification : ∀xPx → ( E!s → Ps ) • Particularization: Ps → ( E!s → ∃xPx ) Lambert’s square sets positive and negative free logics equally apart from classical predicate logic, as both deny the classical version of Specification and Particularization , substituting them by the weaker versions presented above. When it comes to the logical relations between quantified formulae and its corresponding instantiations, positive and negative free logics are closer to each other than to classical first-order systems. 4. The Categories Square In Aristotle’s treatise Categories [1], there is an implicit square different from the most-known diagram implicit in On Interpretation : ‘Socrates is ill’ is the contrary of ‘Socrates is well’. Yet we cannot maintain even here that one statement must always be true and the other must always be false. For, if Socrates really 5 exists, one is true and the other is false. But if Socrates does not exist, both the one and the other are false. To say ‘he is ill’ will be false, and to say ‘he is well’ will be false, if no Socrates so much as exists.